Meta-analysis update of association between dopamine transporter SLC6A3 gene polymorphism and smoking cessation

Introduction

Genetic variation is a significant factor in smoking behavior or smoking cessation treatment. For example, variations in dopamine receptor genes or dopamine transporter genes, which play roles in the dopamine pathway, are significantly implicated in nicotine addiction (David and Munafò, 2008; Stapleton et al., 2007; Ton et al., 2007). Additionally, genetic variants of nicotinic acetylcholine receptor subunits are directly linked to nicotine dependence (Kalamida et al., 2007; Polosa and Benowitz, 2011). Other genetic variations associated with smoking behavior have also been identified, including variations in the catechol-O-methyltransferase and serotonin transporter genes (David and Munafò, 2008; Lerman et al., 1998). Variant alleles of cytochrome P450 2A6 or 2B6 genes are known to play a role in the metabolism of nicotine, which influences smoking cessation and response to treatment (Dicke et al., 2005; Mwenifumbo and Tyndale, 2007; Yamazaki et al., 1999).

Dopamine transporter SLC6A3 gene, also known as DAT1 gene, regulates synaptic dopamine levels by encoding a reuptake protein (Vandenbergh et al., 1992). Genetic variations in a 40-bp variable number of tandem repeats (VNTR) polymorphism in the 3’-untranslated region are related to changes in the density of transporter molecules on the surface of neurons and in dopamine transporter binding (Fuke et al., 2001; Jacobsen et al., 2000). SLC6A3 gene can contain 3–11 repeats, and variation in the number of repeats is associated with idiopathic epilepsy, attention-deficit hyperactivity disorder, dependence on alcohol and cocaine, susceptibility to Parkinson’s disease, and protection against nicotine dependence (Hansen et al., 2014; Jasiewicz et al., 2014).

The 9-repeat allele and 10-repeat allele exist most frequently, and the 9-repeat allele has been associated with reductions in the level of transporter molecules on the surface of neurons (Heinz and Goldman, 2000; Van Dyck et al., 2005). Differing dopamine levels due to changing transporter expression could be related to smoking habits, especially quitting smoking. That is, 9-repeat allele carriers may find it easier to stop smoking than 10-repeat allele carriers (Cook et al., 1995; Gelernter et al., 1994).

Recent studies have evaluated the association between SLC6A3 gene polymorphism and smoking cessation. A meta-analysis by Stapleton et al. (2007) was published previously to review the reported studies. However, since then, more trials have been published, and we conducted this study to present updated results of the meta-analysis.

Materials and methods

Results

Study quality and characteristics

In total, 450 articles were identified through the literature search. After removal of duplicates, the titles and abstracts of 342 articles were screened. Of these, 286 articles were excluded, and the full texts of the remaining 56 articles were assessed for eligibility. A further 47 articles were excluded due to insufficient data, overlapping data, or as review articles. The remaining eight articles were included in this meta-analysis. Figure 1 shows the study selection flow chart according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA; Moher et al., 2009).

OPEN IN VIEWER

Figure 1.

PRISMA diagram of the process of selecting relevant studies.

The characteristics of the eight studies are listed in Table 1. In the quality assessment of the included studies using NOS, four studies were awarded at least five A scores (the highest score for each item) among eight quality score items. Five studies scored six stars and three studies scored four stars.

OPEN IN VIEWER

Table 1.

General characteristics of included studies.

Study	Population	Participants (total or M/F)	Ages (years)	Smoking level (cigarettes/day)	Smoking cessation treatment	Follow-up period	Cessation measures	Conformity with HWE
Sabol et al. (1999)	USA	514	33.7 ± 13.5	Current smokers: 19.9 ± 12.2; Former smokers: 19.0 ± 14.9	No	N/A	Self-reported	N/A
Jorm et al. (2000)	Australia	409	N/A	N/A	No	N/A	Self-reported	N/A
Vandenbergh et al. (2002)	USA	251	45.5 ± 16.9	N/A	No	N/A	Self-reported	Yes
Lerman et al. (2003)	USA	193/225	44 ± 11	25 ± 12	Treatment with bupropion for 10 weeks	6 months	CO measured a	Yes
O’Gara et al. (2007)	UK	225/316	9/9, 9/10 alleles: 43.9 ± 12.4; 10/10 alleles: 43.5 ± 12.9	N/A	Treatment with NRT or bupropion for 6 weeks	10 weeks	CO measured a	Yes
Ton et al. (2007)	USA	0/593	<40 years: n = 203; 40–49 years: n = 222; 50–59 years: n = 119; ⩾60 years: n = 19	N/A	No	6 months	Cotinine measured b	Yes
Han et al. (2008)	South Korea	225/0	AG: 30.0 ± 15.0; NAG: 32.5 ± 13.5	N/A	Treatment with bupropion for 4 weeks	4 weeks	CO measured a	Yes
Styn et al. (2009)	USA	423/458	50–59 years: n = 530; 60–69 years: n = 253; 70–79 years: n = 98	N/A	No	12 months	Self-reported	Yes

AG: abstinence group; NAG: non-abstinence group; HWE: Hardy–Weinberg equilibrium; N/A: not analyzed; NRT: nicotine replacement therapy.

a

Expired-air carbon monoxide < 10 ppm.

b

Cotinine level > 14 mg/mL.

Analysis of SLC6A3 9-repeat alleles versus no 9-repeat alleles

Eight studies were assessed regarding the association between the SLC6A3 gene polymorphism and smoking cessation by comparing two groups: one or two 9-repeat alleles versus no 9-repeat alleles (Figure 2). In total, 678 of 1631 participants (41.6%) with one or two 9-repeat alleles succeeded in smoking cessation, while 830 of 2145 participants (38.7%) with no 9-repeat alleles succeeded. A significant association between the SLC6A3 gene polymorphism and smoking cessation was not observed (odds ratio (OR) = 1.128; 95% confidence interval (CI) = 0.981–1.298). Reanalysis using a random-effects model also showed the same result (OR = 1.904; 95% CI = 0.888–1.348).

OPEN IN VIEWER

Figure 2.

Forest plot of main analysis of SLC6A3 9-repeat alleles versus no 9-repeat alleles.

Subgroup analysis of SLC6A3 9-repeat alleles versus no 9-repeat alleles following smoking cessation treatment

Three studies were assessed regarding the association between SLC6A3 gene polymorphism and smoking cessation with nicotine or bupropion by comparing one or two 9-repeat alleles versus no 9-repeat alleles (Figure 3). In total, 235 of 458 participants (51.3%) with one or two 9-repeat alleles and 351 of 726 participants (48.3%) with no 9-repeat alleles succeeded in smoking cessation. No influence of the SLC6A3 gene polymorphism on smoking cessation treatment with nicotine or bupropion was observed (OR = 1.062; 95% CI = 0.827–1.363). Reanalysis using a random-effects model also showed no significant difference (OR = 0.881; 95% CI = 0.500–1.552).

OPEN IN VIEWER

Figure 3.

Forest plot of subgroup analysis of SLC6A3 9-repeat alleles versus no 9-repeat alleles following smoking cessation treatment.

Sensitivity analyses and publication bias

Sensitivity analysis was conducted by recalculating the results of the meta-analysis after each study was omitted in turn. No considerable differences were observed (data available on request). An assessment of publication bias among all of the available meta-analyses of more than three studies was also conducted. The shapes of the funnel plots revealed no evidence of obvious asymmetry. The results of Begg’s rank correlation test and Egger’s regression test are listed in Table 2. Publication bias was not observed in any of the comparisons.

OPEN IN VIEWER

Table 2.

Test of heterogeneity and publication bias.

	No. of study	Test of heterogeneity			Publication bias
		Q	p value	I 2	p value (Begg’s)	p value (Egger’s)
Main analysis	8	14.2	0.048	50.7	0.711	0.625
Subgroup analysis	3	7.28	0.026	72.5	1.00	0.102

Discussion

This meta-analysis was conducted to review and update the influence of the SLC6A3 gene polymorphism on smoking cessation outcome in terms of abstinence rate. The present meta-analysis quantitatively reviewed the inconsistent results of previous studies and related these findings to the polymorphism.

Prior meta-analysis assessing five studies reported an association between SLC6A3 genotype and smoking cessation such that one or two 9-repeat alleles are related to a greater likelihood of being able to stop smoking (Stapleton et al., 2007). Since then, one clinical trial and two cohort studies additionally evaluated the associations. A prospective cohort study found that increases in the likelihood of quitting in the short term were observed in participants with one or two 9-repeat alleles, but there was no association with long-term cessation (Ton et al., 2007). In contrast, Han et al. (2008) reported a clinical trial in which patients with 10-repeat alleles showed better abstinence rates than those with 9-repeat alleles. Another cohort study suggested that SLC6A3 genotype was not associated with smoking cessation at 1 year (Styn et al., 2009).

In the present meta-analysis, the comparison between 9/9 or 9/* alleles and no 9-repeat alleles of the SLC6A3 gene polymorphism revealed no significant association with smoking cessation outcomes. That is, the genetic variations in SLC6A3 are not significantly associated with smoking cessation outcome, which is inconsistent with the result of previous meta-analysis due to the addition of newly reported studies. Although the gene variants of SLC6A3 are related to various physiologic changes regulating the dopamine pathway, the influence on smoking cessation is not sufficient to present as a clinical result. These results support the notion that clinical applications of pharmacogenetic tests to smoking cessation therapy or counselling are not routinely necessary.

However, the inconsistent conclusions between previous and present meta-analysis studies are very hard to explain simply because cigarette smoking is a very complicated behavior that is influenced by various factors such as age, gender, environment, and race. Another considerable factor is gene–gene interactions that are involved in smoking behavior. For example, one interesting result suggests that a significant interaction between dopamine D2 receptor gene and SLC6A3 gene prolonged smoking cessation and time to relapse (Lerman et al., 2003). Thus, to more accurately account for the association between the SLC6A3 gene polymorphism and smoking cessation, further analyses adjusting for such factors are needed.

In the subgroup analysis following smoking cessation treatment with nicotine or bupropion, associations between the SLC6A3 gene polymorphism and smoking cessation therapy were not observed in the comparisons of 9-repeat alleles and no 9-repeat alleles. Thus, the SLC6A3 genotypes have no correlation to physical mechanisms of smoking cessation therapy or smoking cessation outcomes. The above data are limited in that only three studies were analyzed, but the abstinence rates including only those studies in subgroup analysis were apparently higher than those including all studies. This finding could be explained by treatment effects of nicotine or bupropion on smoking cessation.

Interestingly, several studies with diverse topics about smoking cessation have been reported. For example, Naughton et al. (2015) evaluated the effectiveness of lapse prevention strategies among pregnant smokers. In the results, using self-talk or avoiding spending time with other smokers was a significant help to smoking abstinence. In another example, associations between smoking and depression were suggested from the clinical trial of bupropion in African-American light smokers (Berg et al., 2012).

The present meta-analysis was still limited by the low number and the quality of studies available for inclusion. Additionally, this analysis was based on previously reported studies that were not necessarily complete or accurate. Despite these limitations, our study presents a quantitative approach for integrating updated results in order to increase power.

In conclusion, our study provides newly valid results suggesting that SLC6A3 gene polymorphisms have little to do with smoking cessation outcomes. However, SLC6A3 gene is still an important smoking-related candidate gene. As is well known, recent studies have linked smoking behavior to multiple genes. Thus, further studies measuring combined genetic effects are needed to more accurately account for the influence of genetic polymorphisms on smoking cessation.